ABSTRACT The amount of sensory information available in the world would be overwhelming if we weren?t able to rapidly and flexibly select specific stimuli that are relevant to our current needs. Endogenous control of sensory processing enables improvements in both perceptual sensitivity and behavioral performance and disruption of these cognitive mechanisms in psychiatric diseases have been linked to deficits in behavioral flexibility. Attention to specific stimulus features may make sensory encoding more efficient by modulating activity in cortical networks that process those unique features. Indeed, in visual cortical areas, when attention is directed toward the receptive field of a neuron, the firing rate increases and coordinated activity with other neurons changes. These findings suggest that attention improves perceptual performance by improving the signal carried by specific pools of neurons to their downstream targets for interpreting the visual scene. Additionally, since these changes occur on a rapid timescale, they must be mediated by changes in the functional connectivity that bias feed-forward processing. Determining which visual cortical circuits rapidly change their activity and connectivity with attention will require a behavioral task developed in an organism in which tools for monitoring and manipulating activity in specific subsets of neurons can be used. To this end, I have developed a cued, multi- modal attention paradigm for head-fixed mice; this will allow me to use two-photon calcium imaging to monitor neuronal activity in the primary visual cortex (V1) as the mouse performs the task. In this project I will test the hypothesis that specialized subnetworks of V1 neurons are rapidly and selectively modulated by attention to the features they encode. My preliminary data suggests that V1 neuron responses, on average, increase their activity when visual stimuli are behaviorally relevant. In Aim 1, I will determine the effect and time-course of attentional modulation in mouse V1 to test the hypothesis that V1 neurons are modulated on a rapid, trial-to-trial timescale. Feature attention may improve encoding in V1 by increasing the signal-to-noise of neurons that respond best to those features. In Aim 2, I will identify functional subgroups of neurons based on their tuning properties to the visual stimuli used in the task, and determine which are specifically modulated by attention. In Aim 3, I will determine whether anatomically defined populations are specifically modulated by attention. In these experiments, I will image V1 axon terminals in the higher visual areas, thus isolating specific cortico-cortical projections to compare if these different processing streams change their activity when visual stimuli are attended. This latter experiment will reveal how the effects of attention on visual processing are passed on to downstream targets to affect behavior. By opening new avenues for determining the cellular and circuit mechanisms of attention, this project will bring the field closer to understanding how sensory processing machinery can be optimized to guide perception and behavior.